‘Ceci n’est pas une menace’ : securitizing and desecuritizing discourse in the debate over the Intelligences and Security Services Act 2017 in The Netherlands

(1)

(2)

(3)

1_{Torczyner, Harry (1979). Magritte: Ideas and Images.}

p. 71.

Prologue: the title of the thesis

‘Ceci n’est pas une menace’ is an allusion to the famous painting ‘Ceci n’est pas une pipe’ by the famous Belgian surrealist painter René Magritte. This work of art caused a big controversy over the definition of art and the role of language in representing reality. When asked about this painting, Magritte supposedly remarked:

‘How people reproached me for it! And yet, could you stuff my pipe? No, it's just a representation, is it not? So if I had written on my picture 'This is a pipe', I'd have been lying!’1

The word ‘menace’ is French for ‘threat’ and plays to the emphasis on the role of language in constructing and deconstructing ‘threats’ in the present thesis.

(4)

Chapter 1. Introduction

1.0 Introduction

In February 2017, a new ‘Wet voor Inlichtingen- en Veiligheidsdiensten [Intelligence and Security Services Act]’ was adopted by the Dutch parliament (Anon., 2017). The law dictates what the Dutch intelligence and security services are and describes its entire legal framework (Algemene Inlichtingen- en Veiligheidsdienst, 2017). For half a year, there was very little public attention for the new legislation, but in September of that year five students of the University of Amsterdam organized a petition for an advisory referendum (Anon., 2017). In six weeks, the threshold of 300.000 signatures was reached, resulting in a referendum on the acceptance or rejection of the act held on March 21st_{, 2018 (Van den Dool & Van Lonkhuyzen,} 2017). From this moment onwards, the WIV2017 (which is the official acronym for the new law) was covered extensively by the mainstream media and turned into an issue of widespread public debate. The most controversial element of the revised law became the new power of the general intelligence service (AIVD, Algemene Inlichtingen- en Veiligheidsdienst) and military intelligence service (MIVD, Militaire Inlichtingen- en Veiligheidsdienst) to intercept internet and telephone data from fiber cables instead of radio and satellites only.

Of the 51.6% of the eligible voters that showed up, 46.5 percent voted in favor of the new law and 49.5% voted against the new law, resulting in a negative advise with regards to the WIV2017 towards the Dutch parliament (Kiesraad, 2018). With some alterations, the original law took effect in May 2018 (Niemantsverdriet & Van den Dool, 2018). After this, the opinions were divided on whether the adaptations did justice to the criticisms articulated in the public debate, and some parties already announced to take the new act to the European Court of Justice in order to have it reversed (Hendrickx, 2018). Whether or not the WIV2017 will stay implemented, it is at least remarkable that this law of 66 – rather technical – pages evolved into a big controversy that 7 million people decided to cast their vote on (Kiesraad, 2018).

1.1 State of the art

Because of the recency of the topic, relatively little academic research has been conducted regarding the Intelligence and Security Services Act 2017. A report published in March 2018 by the Dutch Research and Documentation Centre of the Ministry of Justice and Security has thus far been the most comprehensive collection of the new legislation. It is titled ‘The secret services and the democratic rule of law’ and includes a piece on the most important differences

(7)

7 between the new law and the old law (Dielemans, 2018). Mireille Hagens contributed a piece on the issue of oversight over the services (Hagens, 2018), while Nico van Eijk and Quirine Eijkman evaluated the new legislation from a human rights perspective (Van Eijk & Eijkman, 2018). Other authors reflected on the controversy over the act and the place of the Intelligence and Security Services in Dutch democracy (De Graaff & Hijzen, 2018) (Braat, 2018).

One article in the Dutch Research and Documentation Centre issue is of specific relevance to the current paper: Constant Huijzen (2018) reflects on the use of metaphors regarding the Dutch Intelligence and Security Services throughout history. He writes that the Dutch secret services have been compared to ‘ever-growing mushrooms’, ‘barbed wire barrage to democracy’ and ‘James Bond in clogs’ (Huijzen, 2018, p. 5). In his view, it is interesting to see how the recent ‘dragnet’ metaphor causes societal ‘swells’: his study suggests that the use of certain language to describe the new legislation had a profound impact on how it was received by the public. He did not, however, investigate the concrete effects this metaphoric language has had.

The present study follows the perspective of Huijzen, suspecting that the specific use of language profoundly shaped the public understanding of the WIV2017.

1.2 Research questions

The current thesis investigates the debate regarding the WIV2017 in the Dutch political realm and mainstream media. It does so from a Securitization Theory perspective, relating the discourses present in the debate to the concepts of securitization and desecuritization (which will be discussed in the next chapter). Thus, the main research question is:

‘How do the discourses in the WIV2017 debate in the Netherlands relate to the concepts of securitization and desecuritization?’

In order to get a substantiated answer to this question, the following sub-questions will be attempted to answer:

1. ‘Which discourses regarding the ‘Wet op de inlichtingen- en veiligheidsdiensten 2017’ were present in the political realm and mainstream media in The Netherlands?’ 2. ‘How did the discursive struggle over naming the new legislation influence the

(8)

Page | 8

3. ‘How did metaphoric language influence the discourses in the debate?’

4. ‘How do the discourses in the debate relate to securitization and desecuritization?’

As discussed in the previous section, no other research has thus far attempted to study the discourses and use of metaphoric language in the debate over the WIV2017. The present research attempts to provide a deeper insight into the mechanisms behind the dialogical process to which the law was subject. The controversy over the act may be viewed as a unique and exceptional case. However, it may also be viewed as being emblematic for large scale public contestation to the emerging digitalization and mass-surveillance of the public domain. In another vein, it can be seen as a case study of the measureless influence of discourse on the making and unmaking of security, and affect or even create political issues,

1.3 Contents

The current section provides an overview of the content of the thesis per chapter. The next chapter describes the theoretical debates in which the research is situated (chapter 2). After this, a section is dedicated to the research design (chapter 3), followed by two analytical chapters (chapter 4 and 5). The thesis ends with a concluding chapter, discussing the implications of the findings for the academic debate and the possibilities for future research (chapter 6).

As the subject of the debate is a law regarding the scope of powers of the intelligence and security services, the thesis investigates the debate from a security perspective. Chapter 2 describes the theoretical debate in this field, with special attention for Securitization Theory by the Copenhagen School (CS). Although this theory has been very influential in the past two decades, it has also been subject to much criticism on several aspects. Of these criticisms, chapter 2 discusses securitization’s ‘ideal’ counterpart of desecuritization, its reliance on or independence of an ‘audience’, and its struggle over defining ‘extraordinary’ (and, with that, ‘normal’) politics.

As Securitization Theory emphasizes the power and importance of discourses, the current thesis investigates the production and interplay of discourses regarding the new legislation in the Dutch political and public realm. Firstly, chapter 3 describes a poststructuralist discourse analysis in order to study the discursive struggle over naming the

(9)

9 new law. After that, it presents Argumentative Discourse Analysis, based on the work of Maarten Hajer (2006). This method implies to use the conceptual tools of narratives, story lines, and metaphors to study discourse. In order to tailor this method to Securitization Theory, I propose the structuration of the narratives through the concepts of diagnosis, referent object, and prognosis. The chapter then describes the selected sources, including the official stenography of the parliamentary debate regarding the law, articles from the largest four Dutch newspapers, statements by the Dutch general intelligence and security agency, and internet pages of Non-Governmental Organization (NGO) Bits of Freedom. The chapter ends with a section on the limitations of the study.

Chapter 4 presents the analysis of the discursive struggle over naming the law. It discusses the most prominent names in the debate, including the official names and its derivatives, which had a relatively positive connotation. Afterwards, the ‘nicknames’ with negative connotations are examined in detail. The chapter ends with an analysis of the discourse-coalitions promoting the use of some of the names, effectively framing the public debate. Two broad coalitions are presented: the wiv-coalition and the sleepwet-coalition.

Chapter 5 is dedicated to presenting the story lines found in the sources, starting with the pro-side of the debate. Each sub-chapter deals with one story line that may consist of several narratives, starting with ‘the new law is needed against the increased risk of terrorism’, the most prominent story line on the pro-side. The referent object of this narrative is the functioning of the entire ‘Dutch society’, but also its deeper cultural norms and values such as freedom of expression and freedom of religion. Secondly, the story line that ‘the law is needed against increased risk of cyber-attacks and espionage’ is presented. The main narrative presents the cyber space as a vulnerability in the defense of The Netherlands. The referent objects of this narrative are Dutch national sovereignty, its functioning democracy, and the society in general. On the contra-side of the debate, many technical arguments were given that all contributed to the story line ‘the new law compromises fundamental freedoms’. The story line goes that the many weaknesses in the law put the privacy of people in The Netherlands in danger. The referent object of this category is individual, innocent Dutch citizens. The second story line on the contra-side is that ‘the new law is ineffective and/or counterproductive’, which contains two narratives. In the first, the new law is presented to cause an excess of data for the secret services, leading to counterproductivity. Subsequently, the narrative states that the new law decreases the general safety in The Netherlands. The other narrative in this story line states that

(10)

Page | 10

the law simply does not increase the protection against terrorism. Therefore, the law compromises the privacy of Dutch citizens unnecessarily.

The thesis ends with a concluding chapter, stating that the discourse on the pro-side of the debate strongly resemble the classical notion of securitizing speech acts. The narratives on the contra-side, however, do not follow the strategies of desecuritization as proposed by Huysmans (2014). Instead, they are in line with Aradau’s conceptualization of counter-securitization (2004). Yet, the narratives on the oppositional side of the debate had the effect of opening the discussion on what the secret services should and should not be allowed to do. Through this, the public debate also included the dialogue on how the country should deal with an emerging cyber-threat. In the end, the move of this point of discussion from the undemocratic realm of ‘experts’ to the public is exactly what desecuritization entails. Hence, the thesis argues that counter-securitization in some instances might be seen as a strategy for the desecuritization. The thesis ends with a note that a more comprehensive model of the making and unmaking of security is needed, including a clearer conceptualization of the dialogical process of securitizing moves and the contestation thereof.

(11)

11

Chapter 2. Theoretical Framework

2.0 Introduction

The previous chapter showed that very little research has thus far been conducted regarding the discourses in the debate over the WIV2017 due to its recent development. For this exact reason, a sound theoretical foundation is key for studying the course and influence of the debate. The current chapter lays out the theoretical framework guiding the research. Firstly, it describes the relevance of studying the WIV2017 debate through the proposed theoretical lens. The fundament of this lens is Securitization Theory, originally introduced by Ole Wæver, Barry Buzan, and Jaap de Wilde (Buzan, 1997) (Buzan, et al., 1998) . After discussing the main concepts of the theory, the chapter describes the more recent theoretical debates around several criticisms and gaps in the theory. This includes the controversy over the concept of desecuritization, following the argumentation of Claudia Aradau (2004), Columba Peoples and Nick Vaughan-Williams (2014), and Jef Huysmans (2014). The second gap in the literature revolves around the role of the audience in securitization, discussed amongst others by Thierry Balzacq (2005), Noortje Marres (2007) and Paul Roe (2012). The third and last criticism reviewed by this chapter is the issue of defining normal politics, reflecting on an academic dialogue between Aradau (2004) and Roe (2012). This thesis attempts to contribute to these theoretical debates by elucidating to what extent the mechanisms in the WIV2017 debate resonated with securitization and desecuritization.

2.1 Securitization Theory

Securitization Theory was developed by Ole Wæver, Barry Buzan, and Jaap de Wilde between 1995 and 1998 (Buzan, 1997) (Buzan, et al., 1998), and has been a dominant theory in critical security studies ever since. In short, it rejects a classical realist reading of politics, in which the global competition between actors (mostly nation states) for power is central. In study of security, political realism leads to the localization and comparison of measurable threats from competing actors (Peoples & Vaughan-Williams, 2014). Instead, Securitization Theory follows a more constructivist understanding of politics, in which security threats are not as clear-cut and straight-forward, but rather (at least partly) the result of social construction (Buzan, et al., 1998). This school of thought argues that certain political issues shift from the realm of ‘regular’ political debate into the realm of emergency politics through its presentation as an existential threat (Buzan, et al., 1998, p. 21). The referent object this threat is directed at may be a state, society, religion, economy, or any other type of subject. When securitized, the very

(12)

Page | 12

existence of the referent object is presented to be in grave danger, thus justifying extraordinary measures to deal with the threat (Peoples & Vaughan-Williams, 2014, p. 76). The securitizing actors typically perform a securitizing move through a securitizing speech act: by referring to an issue as an issue of security, they create a sense of urgency. With this justification, the actors make a case to apply extraordinary measures so that the referent object can survive (Peoples & Vaughan-Williams, 2014, p. 76). The Copenhagen School (CS), as the original creators of the theory are often referred to, did acknowledge that the context and authority of the securitizing actor have a profound impact on the chance of ‘success’ of the securitization move. This is in line with the sociological idea that no power is ever absolute, but always must be ‘accepted’ to some degree by the people involved. In Securitization Theory, this is called the ‘audience’ (Buzan, et al., 1998). Thus, ultimately, a successful securitization is only possible when the audience accepts it as such. Securitization Theory has a strong constructivist element: the concept of ‘security’ is constructed through utterances of people in places of authority (Peoples & Vaughan-Williams, 2014). Therefore, according to the Copenhagen School, it is not possible to assess how ‘real’ or urgent a threat is, as these variables are dependent on how the threat is dealt with. Accordingly, Securitization Theory has a strong methodological emphasis on the study of discourses, through which certain issues are constructed as issues of security (Mutlu & Salter, 2013, p. 114). Generally, the school of thought treats securitization as a negative trend that is detrimental to democracies, although Wæver (2011) later acknowledged that the urgency and extraordinary measures of securitization are in fact necessary for and beneficial to some issues. In most sectors, however, securitization is regarded a bad development that needs to be constrained, or even reversed. The proposed method to achieve this ‘reverse securitization’ is called desecuritization, as is described in the following section.

2.2 Desecuritization

The process of desecuritization describes how an issue is shifted ‘out of the realm of securitization and emergency politics back into the realm of ‘normal’ political or technical debate’ (Peoples & Vaughan-Williams, 2014, p. 76). Desecuritization has mostly been discussed as the normative element in securitization theory (e.g. Wæver, 2011; Huysmans, 2014). It is often considered the ‘solution’ to the ‘problem’ of the securitization, increasingly observed indifferent domains (Aradau, 2004). However, there are very few instances in which this theoretical idea has been studied in a descriptive manner. As Claudia Aradau (2004, p. 2) notes, desecuritization has received ‘comparatively scant attention’ vis-à-vis the more prominent concept of securitization. While it has been established which concepts to use in

(13)

13 order to study securitization, the theory has thus far been rather ambiguous about the conceptual framework that can be utilized for the analytical study of desecuritization (Peoples & Vaughan-Williams, 2014, p. 85). Jef Huysmans, when investigating the securitization of migration flows in the EU (in which migrants are presented as a threat to existing forms of identity) suggests the existence of three potential strategies of desecuritization:

‘[1] an ‘objectivist strategy’ (we might try to prove that migrants are not really a threat to 'our' identity); [2] a ‘constructivist strategy’ (developing a broader understanding and awareness of how migrants are constructed as threats in processes of securitization as a possible means of undercutting the potency of securitization moves); and [3] a ‘deconstructivist strategy’ (where we might try and listen to the voices and experiences of migrants themselves as means of breaking down exclusionary notions of 'us' and 'them').’

(Huysmans, 1995: 66-67, emphasis added)

These concepts may be used to study the potential desecuritization of cyberspace through the Dutch contestation to the WIV2017. A concept that has gotten very little attention is the concept of counter-securitization, coined by Claudia Aradau (2004, p. 12). This concept describes the securitization of an issue that has already been securitized with regards to a different referent object. Aradau describes this as a normatively bad phenomenon, as she observes that securitization is always at the cost of a group of ‘others’:

‘Who is to be made dangerous so that others be made secure? On which grounds can one privilege such a construction of security, the security of migrants over the security of racists, the security of HIV-positive people over those at risk of being infected?’

(Aradau, 2004, p.12)

While the research in this thesis uses the descriptive concept of counter-securitization, it does not go into the normative question whether one securitization move is more valid than another. It is important, however, to acknowledge Aradau’s resistance to the construction of threats an sich.

2.3 Audience

Apart from desecuritization, another widely criticized aspect of Securitization Theory is the performativity in the process of securitization, which is closely connected to the role of the ‘audience’ (Peoples & Vaughan-Williams, 2014). Balzacq (2005, p. 173) notes that ‘although

(14)

Page | 14

the CS points out that a ‘significant audience’ must concur with the securitizing actor (who speaks ‘security’) for a referent subject, i.e. the threatening event to be securitized–the nature and status of that audience– remains unaccounted for’. Although the Copenhagen School admits that the acceptance of an audience is needed for a ‘successful’ securitization, it does not specify the mechanisms behind this. This conflicting nature of securitization can be described with the terms illocutionary (in which the act is self-referential: something is done in saying ‘security’) and perlocutionary (in which the speech act is conceived as a ‘securitization move’ that only bears performative power when it is accepted by a certain audience) (Roe, 2012). Balzacq describes the tension as follows:

‘[E]ither we argue that securitization is a self-referential practice, in which case we forsake perlocution with the related acquiescence of the audience ... or we hold fast to the creed that using the conception of security also produces a perlocutionary effect, in which case we abandon self-referentiality. I suspect instead that the CS [Copenhagen School] leans towards the first option.... [A]lthough the CS appeals to an audience, its framework ignores the audience, which suggests that the CS opts for an illocutionary view of security yielding a ‘magical efficiency’ rather than a fully-fledged model encompassing perlocution as well.

(Balzacq, 2005: 177–8)

Although it might seem contradictory, the current thesis studies securitization both as a perlocutionary and illocutionary phenomenon, as two different aspects of securitization get to light when using these definitions. Although a word or metaphor might be powerful in itself, for example because it is a tool for the actor to make sense of the world around him, this power can only be asserted on others when it is used in an interpersonal space, and to some extent accepted.

Although the under-conceptualization of the role of audiences in the securitization process is an often-repeated critique, few scholars have attempted to fill this gap in the literature. The issue-oriented conceptualization of ‘audiences’ or ‘the public’ by Noortje Marres (2007) may be a useful to understand the role of audiences better. This approach observes that ‘publics come into being as an effect of changing consequences of human action, which existing institutions can’t accommodate’ (Marres, 2007, p. 769). The study of the debate regarding the WIV2017 fits well with Marres’ notion of an issue-oriented approach. As described in chapter 1, the phenomenon she describes is visible in the topic of this thesis: while there was no such thing as a large ‘public’ contesting the new legislation in February 2017, it was definitely present in Dutch society seven months later.

(15)

15

2.4 ‘Normal’ politics

A third issue that has been widely debated in securitization literature is that of what constitutes ‘normal’ politics. As the call for ‘extraordinary measures’ is an important signifier to study a securitization move, it is important to define what the difference is between ‘ordinary’ and ‘extraordinary’ politics. In the classic securitization literature, it is clear that securitization is seen as a negative development on the basis that it is a result of the ‘failure to deal with issues as normal politics. Ideally, politics should be able to unfold according to routine procedures without this extraordinary elevation of specific threats’ (Buzan et al., 1998: 29). Problematic to this idea is the fact that the Copenhagen School does not explicitly go into the difference between normal politics and extraordinary measures. Implicitly, however, the Copenhagen School relies on an Arendtian understanding of ‘normal’ politics (Roe, 2012). Roe describes this as ‘a space constituted by open discussion between ordinary citizens (together with, or through, their elected representatives) in a bid to solve collective problems’ (2012, p. 251). He adds:

‘Extraordinary politics is, in this sense, what normal politics is not. And normal politics is how things are ordinarily done in liberal democracies. As Aradau (2004: 392) puts it, ‘it is in relation to the procedural “normalcy” of democratic politics that the “exceptionalism” of security can be theorized’. Normal politics thus refers to the notion of ‘routine procedures’. In liberal democratic states, policymaking operates according to established mechanisms. Proposed measures are marked by debate and deliberation. Decision-making is open in the sense that legislatures and other bodies are able to scrutinize the executive.’

(Roe, 2012, p. 251)

In this specific branch of securitization literature, it is hard to distinguish the descriptive from the normative, as (especially in modern liberal democracies) it is difficult to say what ‘normal’ politics is and how ‘normal’ politics should be. Claudia Aradau (2004) defines Securitization Theory as an inherently negative concept. Her two arguments for this are that the extraordinary politics of securitization (1) institutionalizes fast-track decision-making and (2) creates categories of enemy others. These two normative critiques on securitization can be categorized as criticisms directed at its process (the former) and its outcomes (the latter). Roe attempts to refute the first argument by stating that the secrecy, speed, and lack of scrutiny necessary for securitization are heavily overexaggerated in current literature: ‘in the context of liberal democracies, legislation is invariably marked by a greater semblance of oversight than that assumed by Aradau and others’ (Roe, 2012, p. 250). The normative disagreement of these

(16)

Page | 16

scholars about Securitization Theory, however, shows their agreement from a descriptive perspective. Both scholars refer to ‘institutionalized fast-track decision-making’ and ‘secrecy, speed, and lack of scrutiny’. These terms may thus be used as signifiers of extraordinary measures.

(17)

17

Chapter 3. Research Design and Methodology

3.0 Introduction

The previous chapter laid out the theoretical framework in which the research of the thesis is situated. The current chapter describes the research design necessary to investigate which narratives were prominent in the debate surrounding the WIV2017, what the role of metaphoric language was in these narratives, and how they relate to securitization and desecuritization. The following section describes the research methods that follow from the adoption of Securitization Theory and poststructuralism. It argues that discourse analysis is the most useful method to provide insight into the research questions. It specifically discusses the application of Argumentative Discourse Analysis (ADA) as designed by Maarten Hajer (2006), and introduces the conceptual questions guiding the two analytical chapters that follow. For the first analytical chapter, a method with a focus on metaphorical language is proposed to uncover the discursive battle over the naming of the law. Then, some adaptations to Hajer’s method are proposed to uncover the narratives present in the debate in chapter 5. Section 3.2 discusses the selection and collection of data, as well as the coding system applied to the research material collected. Section 4.3 is dedicated to the limitations of the approach taken.

3.1 Methods

Securitization Theory has a strong focus on the construction of social reality through discourse (Peoples & Vaughan-Williams, 2014, p. 78). The focus of the current thesis on the interrelationship between securitization and desecuritization implies to look at the dialogical process in the making and unmaking of security. Therefore, this thesis applies a poststructuralist version of discourse analysis on the debate surrounding the WIV2017, which attempts to study the creation of social reality through the interaction between multiple actors. The core premise of discourse analysis is that ‘language profoundly shapes our view of the world and reality, instead of being merely a neutral medium mirroring it’ (Hajer, 2006, p. 66). Consequently, the definition of a political issue is inherently affected by the particular context in which it is discussed (ibid.). Hence, the analyst taking this approach investigates discourses, here defined as ‘an ensemble of ideas, concepts, and categories through which meaning is given to phenomena and which is produced and reproduced through an identifiable set of practices’ (Hajer, 2006, p. 67).

(18)

Page | 18

Maarten Hajer’s Argumentative Discourse Analysis (ADA, cf. Hajer, 2006) is particularly suited to disclose the discourses articulated by the central actors in the debate surrounding the WIV2017. Its poststructuralist background leads to the focus on the discursive struggle through which meaning is produced. Beste İşleyen describes these competitions of discourse as following:

‘These power struggles are discursive battles ‘that aim to establish a political and moral-intellectual leadership throughout the articulation of meaning and relationship’ (Torfing 2005: 15). The goal is to form a discursive unity with a particular distribution of power in a given social entity.’

(İşleyen, 2014)

Thus, poststructuralist discourse analysis does not view power as ‘coherent and concerted political action’ (Hajer, 2006, p. 69). Instead, it understands politics in terms of many seemingly insignificant micro-mechanisms that impact the interpretation of a certain phenomenon ‘so as to make it manageable for the structures of society’ (ibid.). Thus, through the study of the use of language in a discourse-focused approach, we can attempt to uncover the mechanisms behind these discursive battles. Because social reality is constructed and understood through words, even a single word or name can influence social reality. Robert Chia (2000) calls this ‘organizational discourse’:

‘Our lifeworld is an undifferentiated flux of fleeting sense-impressions and it is out of this brute aboriginal flux of lived experience that attention carves out and conception names ... It is through this process of differentiating, fixing, naming, labelling, classifying and relating - all intrinsic processes of discursive organization - that social reality is systematically constructed.’

(Chia, 2000, p. 1)

The discursive battle over naming the new legislation in mainstream media may thus be seen as an organizational practice. Chapter 4, therefore, analyzes the different names given to the Intelligence and Security Services Act 2017, the connotations these names have, and, thus, how the names influenced the social meaning of the new law.

The study of a discursive struggle in poststructuralist discourse analysis highlights the construction of social reality through interaction. This emphasis on interaction is particularly advantageous for uncovering the interplay between securitization and desecuritization within Securitization Theory. To my knowledge, this thesis is the first attempt in academic literature

(19)

19 to use ADA to study Securitization Theory. The lack of literature using this combination of methods and theory allows for the possibility to uncover important insights in the relations between securitization and desecuritization. As explained below, I propose an adaptation to the original ADA method to make it particularly fitting to study securitization.

ADA uses the concepts of narratives, story lines and metaphors to illuminate distinct features of discourse. Hajer (2006, p. 69) observes that statements are often articulated in a narrative: a story consisting of a beginning, a middle, and an end through which the speaker orders facts or events. In the study of politics, narratives are often not only arranged in a beginning, a middle, and an end, but also in the diagnosis and prognosis of a problem. In the diagnosis, an issue is presented to be a problem for someone or something, which we may call the referent object, a term borrowed from Securitization Theory. In the prognosis, a proposed solution is presented.2 A narrative, thus, provides insight into the assumptions and causal relationships through which the speakers make sense of the world around them. The composition of narratives through a diagnosis, referent object, and prognosis resembles the format of existential threat, referent object, and extraordinary measures belonging to Securitization Theory as described in the previous chapter. The ordering of the narratives as described above thus makes it more convenient to uncover possible cases of securitization. If the diagnosis in a narrative is presented to be of vital importance to the referent object, it is likely to be categorized as an existential threat according to Securitization Theory. Likewise, if the proposed solution in the narrative defies normal politics, the prognosis is likely to fall in the category of extraordinary measures in the framework of Securitization Theory.

Narratives are often rather long and complicated, resulting in actors using a shortened version of them. Hajer (2006, p. 69) calls these story line, defined as ‘condensed statement[s] summarizing complex narratives’. A story line may be accompanied by or composed of a metaphor, as the use of short figurative language is in essence well-suited to capture the meaning of more complex thoughts. In his analysis of metaphors, Huijzen (2018) refers to Lakoff and Johnson’s ‘Metaphors we live by’, arguing that metaphors are not simply fun ways of description, but are a means of understanding and making sense of the world. Metaphoric language (re)structures the conceptual framework of the concerned subject and can thus have real consequences (2018, p. 5). Especially in the topic of the Intelligence and Security Services,

2_{The definition of and causality between the problem, referent object, and prognosis correspond to what Dombos}

(20)

Page | 20

he argues, metaphors have a particular performative power, because people know very little about them due to their secrecy. Metaphors may fill in these blind spots.

Speakers of story lines often implicitly expect their audience to understand the entire narrative by using a shortened statement. However, Hajer, through the application of poststructuralist theory, observes that this presumption of mutual understanding is false more often than not:

‘Very often it is assumed that the meaning that the receiver ‘reads’ in a message is the same as the sender intended to put into the message. This assumption of mutual understanding is false. Discourse analysis brings out, time and again, that people talk at cross-purposes, that people do not really or do not fully understand each other. This is a fact of life but, interestingly, this can be very functional for creating a political coalition.’

(Hajer, 2006, emphasis in original)

Thus, Hajer argues that many political coalitions are based on a false sense of mutual understanding through the common employment of story lines that mean something different for the different users. He calls these political coalitions discourse-coalitions, which he defines as ‘a group of actors that, in the context of an identifiable set of practices, shares the usage of a particular set of story lines over a particular period of time’ (Hajer, 2006, p. 70). As it is clear from this quote, ADA’s object of research is not so much the actors, but rather the ‘story lines’ they utter ‘over a particular period of time’ (idem). Through this discourse-oriented approach, ADA accepts that actors might produce contradictory statements or promote different discourse-coalitions over time (Hajer, 2006).

The figure below shows the conceptual tools discussed above, illustrated with the questions that are applied when studying the selected sources.

Conceptual tool Question

Narrative In what story are the events or facts presented? Diagnosis What is represented to be the problem?

Referent object For what or whom is it represented to be a problem? Prognosis What is the proposed solution?

Story line What condensed statements summarize complex narratives, assuming mutual understanding?

(21)

21 Discourse-coalition Are there groups of actors that, in the context of an identifiable set of

practices, share the usage of a particular set of story lines over a particular period of time?

Finally, in order to fully grasp the political significance of discourses, there is a need to assess their influence on social reality. Hajer (2006) introduces a two-step procedure for measuring this. The first step in this procedure is discourse structuration, which occurs ‘when a discourse starts to dominate the way a given social unit […] conceptualises the world’ (p. 70). This is evident when a new discourse forces central actors to accept its power. The second step that measures the influence of a discourse is discourse institutionalization, when the new discourse is implemented through institutional practices of the applicable political domain.

The following section describes the sources these concepts are applied to. It also describes how these sources were selected and in which manner they were further analyzed.

3.2 Data collection and coding system

In order to discover the dominant narratives in the debate surrounding the WIV2017, key statements by a broad range of actors were analyzed. Especially the study of the formation of discourse-coalitions and the circulation of metaphors and story lines requires the investigation of the interaction amongst multiple actors. Statements were classified as ‘key statement’ on the basis of two things. Firstly, if the statement or medium allowed the actor to reach a large audience, the statement was included. Secondly, if the content or presentation of a statement resonated in many other observed statements, it was included in the selected data. The current section describes the documents that were selected and their relevance for the research.

Firstly, the official stenography of the parliamentary debate regarding the final version of the WIV 2017, on February 8th, 2017, was obtained through the website of the Dutch parliament (Tweede Kamer der Staten Generaal & The Netherlands, 2017). Although the process of shaping and reshaping the law started in 2013, due to time and space constraints it was not possible to include this process in the thesis. Moreover, it is difficult to determine at what point in time the debate over the law started while it was still being written. From the moment the law got the official title ‘Wet op de Inlichtingen- en Veiligheidsdiensten’, there was a concrete and delineated subject of speech. The debate was scheduled to last three to four hours, but instead lasted for nine (Tweede Kamer der Staten Generaal & The Netherlands,

(22)

Page | 22

2017). This shows a first impression of the significance of the issue in the national political realm.

Secondly, the text of the original website of the petition with the call for a referendum by the five students of the University of Amsterdam (Anon., 2017) was analyzed. This website linked to a page of NGO Bits of Freedom with an explanation about the law (Bits of Freedom, 2017). These documents are specifically relevant as they played an important role in the spread of the debate from the political elite to the public (Van den Dool & Van Lonkhuyzen, 2017). Thus, it is reasonable to assume that these texts influenced the way the law was conceptualized by the mainstream media and the public.

Thirdly, a large selection of newspaper articles was collected, as newspaper articles were expected to give an accurate representation of the spread of narratives and metaphors over time. Additionally, in the framework of Securitization Theory, newspapers can be regarded both as an actor (making statements, influencing the audience) and an audience (promoting and rejecting statements made by others). For this, the website LexisNexis was used, on which all Dutch newspaper articles are stored for academic purposes. The four biggest national newspapers were used: De Telegraaf (regarded populist, right-wing aligned, and sensationalist); Trouw (neutral); De Volkskrant (originally Catholic, regarded left-wing); and NRC Handelsblad (progressive, left-wing) (Anon., 2016). These newspapers together have a circulation of over 1.5 million copies and are owned by three different publishers. In order to get a full representation of newspaper articles referring to the law, all articles that included the official title or one of its ‘nicknames’ were gathered through the search function of LexisNexis. The following names were discerned in an inductive search: ‘WIV2017’ or ‘WIV’ [the acronym for the official law]; ‘anti-terreurwet’ [anti-terror law]; ‘tapwet’ or ‘aftapwet’ [tap law]; and ‘sleepwet’ [dragnet law]. The period between the parliamentary debate (February 2017) and the referendum vote on the law (March 2018) was used as the time frame of the search. In this period, the debate spread from the political to the public realm, creating interactions on the issue between actors throughout Dutch society. In total 103 articles were collected. The analytical chapters make general statements about trends observed in the collected newspapers. When direct quotes are used, they are usually representative of these trends, unless noted otherwise.

Fourthly, the transcript of a video from the talk show ‘Zondag met Lubach [Sunday with Lubach]’ was selected (Zondag met Lubach, 2017). On October 1st_{, 2017, Arjen Lubach}

(23)

23 used his popular satire news show to talk about the WIV2017 and the petition to organize a referendum. The video gathered hundreds of thousands of views on multiple platforms (e.g. Youtube.com and Lubach’s website), causing the petition to gain much traction (Van Lonkhuyzen, 2017a). Several sources agree that it was this video that helped the petition pass the mark of 300.000 signatures, i.e. the number necessary for the organization of a referendum (e.g. Van Lonkhuyzen, 2017a; Breebaart, 2017). Hence, this video is of particular relevance for the study of the desecuritization of the issue.

Lastly, several statements of the AIVD were selected. In November 2017, the agency uploaded a section on its website containing an explanation of what the WIV2017 entails, why it is necessary, and answers to frequently asked questions (Algemene Inlichtingen- en Veiligheidsdienst, 2017). These pages were published in order to provide Dutch citizens with the information necessary to make a decision regarding their vote in the referendum. Additionally, a transcript of the appearance of Rob Bertholee (director of the AIVD) in the popular talk show ‘De Wereld Draait Door’ was collected (De Wereld Draait Door, 2018). One month before the referendum vote, he appeared in this popular talk show to present his vision on the new legislation. On average, this news show had 1.7 million viewers per day in the month of February (Stichting Kijkonderzoek, 2018). Thus, on a population of around 17 million people, Bertholee had a very large platform at his disposal to promote his perspective. The texts mentioned above were analyzed using the computer program Atlas.ti, through a combination of inductive and deductive analysis. Firstly, the documents were fully scanned and coded with respect to discussion topics and discursive patterns. Then, drawing on the conceptual framework sketched above (cf. section 3.1), the texts were coded again, using the labels ‘diagnosis’, ‘referent object’, ‘prognosis’, and ‘metaphor’. For the study of the discursive struggle around the naming of the law, all titles and nicknames of the law were coded. These labels were then grouped according to the narrative they were presented in. These narratives were, again, grouped in order to get to the story lines presented in chapter 5.

3.3 Limitations

While the theoretical and conceptual background of the current thesis necessitate the selection of the selection of a diverse set of actors, this limits the study in two ways. Firstly, due to time and space constraints, only a small selection of sources could be selected per actor. This brings into question the representativeness of the wider debate, as some important statements might

(24)

Page | 24

have been overlooked. Secondly, only a limited number of practices could be studied. The research attempted to include a wide set of discursive practices, including a parliamentary debate, online statements on websites, transcriptions of TV speeches and interviews, and news articles from several newspapers. Yet, this still excluded some practices. For example, Lene Hansen (2011) describes the significance of studying the visual aspect of discourse, as this can bear important messages of securitization that would be otherwise overlooked. Finally, although many different sources were taken into account and the newspaper articles aided to get a full view of the actors that were important to the debate, there is a real chance that some important actors were overlooked.

A second limit to this study follows from the adaptation of the ADA method. In this chapter I argued that the structuring of the found narratives through diagnosis, referent object and prognosis is particularly suited to uncover securitizing narratives. A drawback to this is that this methodological lens may therefore be unsuited to uncover other mechanisms in the narratives. In other words, the use of this method detects securitizing narratives because it was designed to do so, but it is unable to compare this discursive mechanism to others. This limitation resonates with Huysmans’ (2002) articulation of the ‘normative dilemma’ of Securitization Studies: by writing through a securitization lens, does a scholar contribute to the notion that everything gets viewed as an issue of security? The current study attempts to balance out this unease through an additional focus on desecuritization and large-scale contestation to securitization processes, but does, to a certain extent, contribute to Huysman’s dilemma.

Another difficulty that is specific to the current study is the fact that the thesis – and, with that, the analysis – is presented in English, while most of the sources described are in Dutch. Because of the nature of the methods of the research, many quotations are used in the analytical chapters. I chose to present an English translation of the quotations in the texts, while presenting the original Dutch statements in footnotes. Especially in chapter 4, which regularly dissects words in order to get a deeper understanding of their connotations, this way of presenting the results proved to produce a more coherent and flowing analysis. In contrast, the in-text inclusion of the Dutch quotes would have resulted in frequent codeswitching, ultimately leading to disorganized analytical chapters. It is important to acknowledge, however, that the translation of the original statement inherently changed some of the original connotations, and, consequently, influenced the analytical process. I took this limitation into account to my best abilities, by sometimes choosing for a more literal translation (at the cost of the flow of the

(25)

25 sentence), while at other times using a slightly divergent translation in order to resemble the original connotation more.

(26)

Page | 26

Chapter 4. Analysis: Discursive Struggle

4.0 Introduction

The previous chapter set out the methodological framework of the thesis, showing that the naming of the new legislation has a discursively organizational nature. The official name of the new law is Wet op de inlichtingen- en veiligheidsdiensten 2017 [Intelligence and Security Services Act 2017]. As it is too long to be comfortably and systematically mentioned in a debate or in a media article, it is unsurprising that the actors involved started using abbreviations and ‘nicknames’ to refer to the proposed legislation. A myriad of nicknames became popular amongst different actors in different periods, to the point where the law was satirically called ‘de wet met de vele namen [the law with the many names]’ (Zondag met Lubach, 2018). These names, however, do not reflect the content of a fairly complex legal instrument in a comprehensive or neutral way. Instead, the nicknames inherently essentialize or reduce the law to its ‘core’ point, which is dependent on the point of view of the actor. Thus, a certain name might both consciously and sub-consciously be used by an actor to convey a message or connotation. If a coalition of actors all use the same name, it automatically redefines the debate, as a certain aspect of the law gets emphasized. The most prominent names are discussed, starting with the use of the official name, ‘Wet op de inlichtingen- en veiligheidsdiensten’ and its acronyms ‘WIV’ and ‘WIV2017’. Then, the use of the abbreviation ‘inlichtingenwet [intelligence law]’ will be discussed. Thirdly, the use of the nicknames ‘tapwet [tap law]’ and ‘aftapwet [tap law]’ will be illustrated. Lastly, the most-used nickname, ‘sleepwet [dragnet law]’ will be explained through a description of the metaphor it was derived from, and its negative connotation. The chapter ends with an analysis of the discourse-coalitions promoting the use of some of the names, effectively framing the public debate. Two broad coalitions are presented: the wiv-coalition and the sleepwet-coalition. These two coalitions strongly influenced the narratives presented in the next chapter.

4.1: Wet op de Inlichtingen- en Veiligheidsdiensten [Intelligence and Security

Services Act], WIV2017 and WIV

The full name is used mainly in official documents. For instance, ‘Wet op de inlichtingen- en veiligheidsdiensten’ is the official title of the proposed legislation as used in the correspondence of the house of representatives. As it is a revised version of the original law, created in 2002, it has the same name, thus often mentioned together with the year of creation

(27)

27 in order to reduce the risk of confusion. In an attempt to use an official and neutral tone, actors trying to use a shorter, catchier name abbreviated it to ‘Wiv’ or ‘WIV2017’. The fact that the name is used in official documents does not mean that it is free of connotations or implications. At least two things are implied by the fact that a legal framework is built around the intelligence and security agencies. Firstly, it implies that the agencies need this framework to demarcate for which tasks they are responsible. However, the fact that most of the activities of the services are conducted in secrecy suggests that the agencies also need the legal framework as a system of checks and balances to not abuse their powers. Thus, the legislation hints at some level of justified suspicion towards the agencies. Rob Bertholée, director of the AIVD, was aware of this in his appearance in the media:

‘I really hope that people realize that this law does not exist to keep a crooked regime in charge, but that the law exists to guarantee the safety of The Netherlands and the Dutch.’3

(Bertholée, 2018)

In this quote, Bertholée tries to profile his service as one that deserves trust, by contrasting the service with those that have supported ‘crooked regimes’. Instead, he emphasizes the role of the agency in looking after the safety of the country and its people. Secondly, the use of this name emphasizes not what the agencies do or have done, but what they will be allowed to do, including in the most extreme hypothetical situations. This incites actors to use examples of (sometimes extreme) hypothetical narratives in order to convince their audience of their proposed extraordinary measures.

In the entire nine hours of the parliamentary debate on February 8th, 2017, the full name was only used seven times to refer to the new legislation, by politicians with different affiliations. The abbreviation ‘Wiv’ was used even more scarcely, only by representatives of parties in favor of the new law. However, because the WIV2017 was the main subject of the parliamentary debate, most politicians simply referred to it as ‘the law’ or ‘this law’.

Additionally, the official name of the law and its abbreviation were used in many of the articles examined in Trouw, NRC, and De Volkskrant. However, these names appeared in the headlines only very rarely. De Telegraaf mentioned the full name or the abbreviations a handful of time, instead referring to the many nicknames. The context in which the official name and abbreviation were used in the newspapers differed considerably. Throughout October and

3_{Original: ‘Maar ik hoop echt dat mensen zich realiseren dat deze wet er niet is om een schimmig regime in het}

(28)

Page | 28

November 2017, the Trouw and the NRC Handelsblad made mention of the official name only in combination with the ‘sleepwet’ name. In February and March 2018, nearing the referendum, these newspapers used the official name more consistently, giving less attention to the nicknames. The Volkskrant showed a reverse trend, mentioning the official name many times in October and November 2017, but referring mostly to the nicknames in February and March 2018.

4.2: Inlichtingenwet [intelligence law]

The name ‘intlichtingenwet’ translates to ‘intelligence law’ and is somewhat of a mixture between an abbreviation and a nickname. The name covers the original meaning of the name of the law relatively well, and thus the use of this name may be interpreted as an attempt to give neutral coverage of the law. It can be argued, however, that the word ‘inlichtingen’ sounds relatively positive, and thus has a constructive connotation to it.

The name ‘inlichtingenwet’ was used very minimally during the newspaper coverage of the law in 2017. In February and March 2018, however, all analyzed newspapers turned to this name as the most common way to refer to the new law. Especially the NRC Handelsblad used this name systematically during that period. An explanation for this could be that, while the referendum was approaching, most newspapers expected that their audiences wanted to base their decision of their vote on articles that at least gave the impression of attempting to be neutral in the matter.

4.3: (Af)tapwet [tap law]

The name ‘aftapwet [tap law]’ was already introduced by the NRC Handelsblad in October 2016, during the time at which the law was being written. Of the four analyzed newspapers only the Trouw and the NRC Handelsblad covered the law before February 2017, and both referred to it as the ‘aftapwet’, using quotation marks. This nickname and its shorter version ‘tapwet [tap law]’ stayed relatively popular until September 2017, when the petition for the referendum gained traction. From this moment onwards, these nicknames lost popularity, only appearing in the analyzed newspapers a handful of times.

(29)

29 The word ‘aftappen [tapping]’ is a synonym for the word ‘interception’, putting emphasis on this aspect of the law (as opposed to, for example, the clauses on hacking or sharing data with foreign parties).

4.4: Sleepnet [dragnet] and sleepwet [dragnet law]

The ‘sleepwet’ name was by far the most used name for the new legislation throughout the studied period. The current section first discusses the use of the word ‘sleepnet’, from which the name is derived. Afterwards, it will zoom out to study the use of the ‘sleepwet’ name by the different actors.

´Sleepnet [dragnet]´

During the parliamentary debate regarding the new law in February 2017, the word ‘sleepnet [dragnet]’ was used frequently (61 times in seven hours) (Tweede Kamer der Staten Generaal, 2017). This word has powerful connotations through imagery. The vehicle of this metaphor, the dragnet, may be used to gather a large sum of fish, as opposed to fishing with a rod. The quality of the total catch, however, is worse than when a rod and bait are used to target specific species. Thus, the dragnet harms many untargeted animals, damaging the ecosystem in the process. The tenor of the metaphor refers to the specific warrant in the law that allows the secret services to gather large amounts of data from fiber cables, including that of many innocent people. As is clear from the vehicle, this metaphor has strong negative undertones: as the dragnet fishing method hurts untargeted fish and damages the ecosystem, it is implied that the dragnet interception method hurts innocent people and damages society.

The negative connotation of the metaphor was used extensively by the opposition in the parliamentary debate. Verhoeven, representative of D66, said in the parliamentary debate of February 2017: ‘The cabinet calls that ‘research assignment-oriented’ in the law. Colloquially that is called a dragnet.’4_{(Verhoeven, 2017, p.2) Not only is the dragnet metaphor} a lot shorter (and, arguably, catchier) than the official name of the proposed authorization, but Verhoeven’s choice for the phrase ‘in de volksmond’ tactically played to the sentiment of his constituency. The translation of this phrase resembles ‘colloquially’ or ‘popularly’, but literally means ‘in the mouth of the people’. The word ‘volk [people]’ has a connotation of unity, so

(30)

Page | 30

with his statement Verhoeven hints at a unified resistance of the ‘normal’ Dutch people towards this aspect of the new law.

This tactic was used systematically by representatives of parties opposing the new law, attempting to give a negative connotation. However, even proponents of the new law could not seem to avoid mentioning the metaphor or playing into it. Tellegen, spokesperson of the VVD, said, for example: ‘In the entire trio of precautionary measures I see enough guarantees, which is why I do not recognize myself in the image – I am not going to use the word – of the broad dragging in of data.’5_{(Tellegen, 2017, p.35) Her quote shows that she was aware of the negative} undertones of the word ‘dragnet’ and explicitly avoided using it. Yet, Tellegen played into the metaphor by referring to the proposed interception as ‘the dragging in of data’. Although her statement was of a dismissive nature, her choice of words still reflected the narrative that this new competence in the law is similar to that of using a dragnet to fish and might thus be harmful. In line with this, several politicians on both the pro and contra side referred to gathered information of innocent people as ‘bijvangst [bycatch]’: ‘Could the minister […] sketch [...] how bycatch will be removed immediately after interception?’6 (Recourt, 2017, p.22). The reference to fishing in the use of the word ‘bycatch’ is clearly an extension of the dragnet metaphor, even though Recourt (spokesperson of the governing Labour Party) avoids the metaphor itself.

As soon as the issue gained traction in the media, the dragnet metaphor got extended in order to describe other points of criticism towards the law:

‘And that part of the wiv is also called the dragnet. If you happen to live in the same neighborhood that some foreign fighters to Syria also come from, then the AIVD is allowed to throw out a digital net, and chances are that your internet traffic also gets fished up, […] so there will be digital dragnets and the secret service only has to throw back the innocent fish after three years.’7

(Lubach, 2017, emphasis added)

5_{Original: ‘In die hele trits aan voorzorgsmaatregelen zie ik genoeg waarborgen waardoor ik mij niet herken in het}

beeld — ik ga het woord niet gebruiken — van dat breed binnenslepen van data.’

6_{Original: ‘Kan de minister [...] schetsen [...] hoe bijvangst na interceptie direct zal worden verwijderd?’}

7_{Original: ‘En dat onderdeel van de wiv wordt wel het sleepnet genoemd. Als je toevallig in dezelfde wijk woont waar}

toevallig ook wat syriegangers vandaan komen, dan mag de AIVD een digitaal net uitgooien, en bestaat de kans dat ook jouw internetverkeer wordt opgevist, [...] er komen dus digitale sleepnetten en de geheime dienst hoeft onschuldige vissen pas na drie jaar terug te gooien.’

(31)

31 In this quote, Lubach used the dragnet metaphor extensively. In line with the original metaphor, he spoke of the ‘dragnet’ part of the law, allowing the secret services to throw out a ‘digital dragnet’. Additionally, he mentioned that ‘your’ data might be ‘fished up’ too, making it a personal and individual matter by addressing his audience directly. Finally, he emphasized the breach of privacy of the ‘normal’ citizens by calling them ‘innocent fish’.

Onderzoeksopdrachtgerichte interceptie [research assignment-oriented interception]

As mentioned above, the official name for the ‘dragnet’ clause of the law is ‘onderzoeksopdrachtgerichte interceptie [research assignment-oriented interception]’. This term was used around 25 times in the parliamentary debate of February 2017, but mainly by opponents of the new law. For example, Ronald van Raak (spokesperson of the SP) said: ‘If a fisher discovers in the future that he is not allowed to use a dragnet, I would advise him to call it a “research assignment-oriented net”. Then he will probably get away with it.’8_{(Van Raak,} 2017, p.69) In this quote, Van Raak turned the popular dragnet metaphor upside down. He mocked the cabinet by implicitly comparing it with the fisher in his story, sending the message that the government gave a new, long, and complicated name to something that they are not allowed to use. In this quote (and many others), the opposition tried to demonstrate that the language of the new legislation is deliberately misleading, effectively maneuvering a competence into the law that the general public would otherwise not tolerate. Regardless of whether Van Raak was right about the notion that the public would not accept this aspect of the law, his metaphor surely touched on something legitimate. By using the long, official name, the cabinet not only gave the impression that the interception was ‘directed’, which is debatable. It also presented the issue as a technical one that is not very comprehensive for the general public, legitimizing this issue to be handled by ‘experts’. This resonates with Securitization Theory as discussed in chapter 2. By presenting the specifics of the interception of digital communication as a technical and expert decision, the Dutch cabinet may move this topic from the politicized level to a depoliticized, securitized one.

Sleepwet [dragnet law]

The name ‘sleepwet’ is a contraction between the Dutch words ‘sleepnet [dragnet]’ and ‘wet [law]’. The closest translation is simply ‘dragnet law’. However, it is important to realize that the name is not only a metaphor, carrying the same connotations as the ‘sleepnet [dragnet]’

8_{Original: ‘Als een visser in de toekomst te horen krijgt dat hij geen sleepnet mag gebruiken, zou ik hem adviseren}

(32)

Page | 32

metaphor. It also is an entirely new word in the Dutch language, specifically constructed for the debate regarding the WIV2017. This means that, when used in the media, there is no doubt as to what the metaphor refers to, because it has never been used to describe anything else. This arguably gives the metaphor extra discursive power and probably contributed substantially to its popularity. The ‘sleepwet’ name was coined first by the group of students organizing the petition for the referendum on their website in September 2017 and was taken over by the media quickly after. All analyzed newspapers started by putting the name between quotation marks and adding nuances such as ‘the Wiv, called ‘dragnet law’ by its opponents…’ and ‘the so-called ‘dragnet law’…’. However, after the nickname had gotten significant popularity, the newspapers used the ‘sleepwet’ name with less subtlety. In most analyzed articles many names for the law were used, as exemplified in an NRC Handelsblad article from November 1st, 2017: ‘There will definitely be a referendum on the new Intelligence and Security Services Act (Wiv). [...] Furthermore, the implementation of this ‘dragnet law’ will be postponed.’9 (NRC Handelsblad, 2017) In this passage, the writer of the article clearly attempted to stay as neutral as possible, first calling the law by its official name, then mentioning the acronym. When the law is referred to for the second time it is called by its most popular nickname, but with quotation marks to show that the name is not official and was created by someone else. Later in the same article, however, the following was stated: ‘Proponents of the dragnet law argued earlier that the law would take effect on January 1st, making it difficult or unnecessary to hold a referendum about it afterwards.’10_{(Ibid.) In this sentence, the quotation marks were omitted,} and defenders of the law were simply called ‘dragnet law proponents’, putting them in a rather negative light.

The websites of the petition organizers and Bits of Freedom consistently kept referring to the ‘sleepwet’ name throughout the run-up to the referendum, but all analyzed newspapers slowly adopted a more nuanced tone, mentioning the official name more often (Trouw and NRC Handelsblad) or including more of the other nicknames (Telegraaf and Volkskrant). This shows that, while the NGO and the organizers of the petition attempted to swing peoples’ opinion on the legislation explicitly, the newspapers attempted to give a more balanced and neutral impression to their audience.

9_{Original: ‘Er komt definitief een referendum over de nieuwe Wet op de inlichtingen- en veiligheidsdiensten (Wiv).}

[...] Ook wordt de invoering van deze ‘sleepwet’ uitgesteld.’

10_{Original: ‘Voorstanders van de sleepwet voerden eerder aan dat de wet op 1 januari in zou gaan en het daarom}

(33)

33 Users of the ‘sleepwet’ name consciously or subconsciously convey two strong discursive messages. The first is that the most important aspect of the new law, and thus the essence of the debate, is the new competence of the secret services to gather cable data in a broad fashion. By uttering this name time and again, actors automatically draw most of the attention in the debate to this specific aspect, marginalizing the significance of other features. The second is that this is a negative development. This message lies in the nature of the dragnet metaphor. As described above, the metaphor has a negative connotation, specifically with regards to ‘innocent fish’ and the effects of the law on the functioning of society.

4.5: Discourse-coalitions

The previous subchapters discussed the use of several names for the WIV2017. The passages argued that the different names given to the law all carry their own connotations, putting the new law in a certain (positive or negative) light, and increasing the salience of certain aspects of it. This section assesses if and how a coalition of powerful actors adopted the use of a certain name, thereby framing or nudging the general debate. The current passage describes the use of certain names by the actors involved, the circulation of the names and metaphors, and the discourse-coalitions that took shape.

In the political realm, where the debate surrounding the law started, the proponents of the law clearly tried to stay away from using any other name for the law than its official name or abbreviation. Speakers from the opposition criticized the law on many aspects, but many of them succeeded in emphasizing the one clause stating that the secret services would be allowed to intercept cable data on a large scale. They did this not only by bringing up the issue time and again, but also by the discursive tactic of consistently comparing this authorization with a dragnet. This discourse-coalition, consisting in the analyzed parliamentary debate most notably of Verhoeven (D66), Voortman (GroenLinks) and Van Raak (SP), was so convincing that the proponents of the law in the parliament were forced to use comparable language and play into the metaphor. This resonates with Hajer’s first signifier of discourse having influence on social reality, which he calls ‘discourse structuration’ (2006, p. 70). Even though the proponents of the law consciously attempted to steer clear of the ‘dragnet’ language, they were forced to accept its discursive power. This discourse-coalition got strengthened half a year later by the organizers of the petition, supported by Bits of Freedom.